Talk:Task parallelism

I think the example section could use a little work, but I'm not an expert enough to say whether it's actually wrong or not so I've left it the way it is. The problem I see is in the pseudocode: Daniel Lee (talk) 08:55, 25 March 2013 (UTC)
 * According to the example, "a" is executed as the first part of the if-block. "b" is part of the same block and is executed if the "else if" statement is true.
 * In my opinion, "b" would never be executed because "a" evaluates as true, so the rest of the statements in that if-block are skipped. All the languages I'm familiar with evaluate each if-statement until they find one that's true and then skip the remaining else-statements, if any. Or is this ideosyncratic? AFAIK, C, C++, Python, Java and Scala all follow this convention.

Task parallelism = MIMD? --Abdull (talk) 09:00, 24 September 2009 (UTC)

The JVM example is confusing, and looks like spam. Furthermore, it adds little to the content of the article. — Preceding unsigned comment added by 85.247.160.254 (talk) 20:16, 6 November 2013 (UTC)

Contrast with data parallelism
I'd don't really see the promised contrast with data parallelism materializing here. It looks like data parallelism is a special case of task parallelism (different data, same code). I'm not sure if that's intended. Q VVERTYVS (hm?) 22:13, 28 December 2014 (UTC)

thread-level parallelism
I see that thread-level parallelism is now a redirect to this task parallelism article.

In another article, I see that someone recently changed every instance of "thread-level parallelism" (TLP) to "task-level parallelism" (TLP).

Is there a difference between "thread-level parallelism" vs "task-level parallelism"? If so, what is the difference? --DavidCary (talk) 16:45, 19 November 2015 (UTC)